Using SPE and HPLC-MS to Quantify and Identify Pharmaceutical Compounds in St. John's University Wastewater

____________________________________________________________________________________ Pharmaceuticals in wastewater have become a concern of environmental toxicologists. An efficient method of discovering the concentrations of these pharmaceuticals in wastewater has not yet been produced. The method we developed includes an automated Solid Phase Extraction (SPE) procedure prior to injecting a sample of wastewater into the High Performance Liquid Chromatograph (HPLC) and Mass Spectrometer Electrospray Ionization (MS-­‐ESI). Three unknown peaks were identified on the HPLC Methods of analysis including NMR, GC-­‐MS and IR have been used to determine the composition of these compounds that are potentially in significant concentration in the wastewater. _____________________________________________________________________________________ INTRODUCTION Until recently, the presence of pharmaceuticals in wastewater has not been a concern of environmental toxicologists. However, studies have shown that concentrations as low as 1 ppb (part per billion) and sometimes 1 ppt (part per trillion) can have adverse environmental effects. Pharmaceuticals can easily be deposited into aquatic environments through effluents such as wastewater, and there is little known about their possible synergistic effects. Because of the potential for biological consequences in various communities, including CSB|SJU, it is critical to determine an efficient method of discovering the concentrations of these pharmaceuticals in wastewater. Current methods of evaluation include manual SPE coupled with LC/MS-­‐ESI(+) and continuous liquid-­‐liquid extraction (CLLE). The method we developed includes an automated, rather than manual, SPE procedure prior to injecting a pre-­‐concentrated sample of wastewater into the HPLC/MS-­‐ESI. This paper describes the research from June 2012 to April 2014 to determine the limit of detection (LOD) of this method starting with known amounts of antidepressants in E-­‐ University, Collegeville, MN. METHODS Part 1: Determining the LOD Preparation of the antidepressant solutions Six antidepressant drugs were dissolved in methanol in various ways to be prepared into approximately 10 mM solutions. Paroxetine HCl (Paxil) was weighed out as a pure substance and added to a 100 mL volumetric flask. Tablets of Sertraline HCl (Zoloft), Quetiapine Fumerate (Geodon), and Escitalopram Oxalate (Lexapro) were crushed with mortar and pistol, transferred into a 100 mL volumetric flask, sonicated for 15 minutes with a Branson 2510 Ultrasonic Cleaner, and filtered using 0.45um nylon filter paper. Similarly, Aripiprazole (Abilify) was crushed, transferred into a 50 mL volumetric flask because there was a limited supply of aripiprazole, sonicated, and filtered. Ziprasidone HCl came as a capsule, and the insides were transferred into the 100 mL volumetric flask, which was sonicated and filtered. Each solution was placed in a separate amber bottle to prevent possible photodecomposition. Direct Injection HPLC One hundred HPLC vials, and 1 mL of E-­‐pure H2O was added to each vial. Two direct injection HPLC methods were attempted with 1.00 of each of these six solutions that did not show any peaks, including a mobile phase of 50% methanol/50% pH 9 Kinetex analytical column with the Thermoscientific Surveyor HPLC at 0.500 mL/min. Every HPLC method used a wavelength of 215 nm. An 80% methanol 20% pH 9 mobile phase run at 1.00 mL/min showed peaks, but they were very close together. At a flow rate of 0.500 mL/min, every peak showed up except for aripiprazole. A mobile phase of 70% methanol 30% pH 9 buffer at a flow rate of 0.500 mL/min showed consistent, distinct peaks for every compound. The LC/MS was added to give MS chromatograms of the data as well. The six antidepressant solutions of about 10 mM were then added together by placing 10 mL of each into one amber bottle. When run under the 70% methanol/30% pH 9 buffer, 0.500 mL/min flow rate conditions, they showed distinct peaks. When precipitate was found in some of the sample vials containing quetiapine fumarate, escitalopram oxalate, and ziprasidone, these three solutions were re-­‐ made using acetonitrile as the solvent instead of methanol, but precipitate formed in the acetonitrile solutions as well. Equal amounts of the 10 mM solutions of paroxetine, sertraline, and aripiprazole were then combined in an amber bottle. The paroxetine, sertraline, aripiprazole (PSA) solution was analyzed under the 70% methanol/30% pH 9 buffer conditions, and then made more dilute until the peaks no longer showed in order to determine the LOD. This method was accomplished by placing 500, 300, 200, into separate amber vials, adding 1 mL of methanol to each, and running it through the 70% methanol/30% pH 9 buffer HPLC method. Manual SPE/HPLC The next step in our process was to make up solutions that were at even lower concentrations than the direct injection HPLC method could detect and do a Solid Phase Extraction (SPE) to concentrate the compound prior to HPLC injection. Doing this iteratively with smaller concentrations each time helped to determine how low of concentrations the HPLC can detect after using SPE. The manual SPE experimental procedure is as follows: 1 L of E-­‐pure water was placed in each of three 3 L amber bottles and then ethanolamine, acetic acid, and formic acid were weighed out so that each jug contained 10 mM of the respective buffer. The pH of each bottle was adjusted to pH 9, pH 5, and pH 3, respectively, with ammonium hydroxide (14.8 M) or HCl (6 M). The LOD of this method was determined by adding the paroxetine, sertraline, aripiprazole (PSA) solution to the adjusted buffer solutions in smaller and smaller amounts. These solutions were then extracted under vacuum at about 15 mmHg using Oasis HLB 6cc SPE columns at a flow rate of 2 mL/min. The columns were prepped with 5 mL 90% TBME/10% MeOH until dried, 5 mL 100% MeOH until dried, and 5 mL 100% E-­‐pure H2O until there remained a thin layer of water on the column. After the columns were loaded with the PSA/buffer solutions, a centrifuge tube was placed underneath the columns and 3-­‐5 mL of 90%TBME/10% MeOH was used to extract the antidepressants from the column. The centrifuge tubes, now containing 3-­‐5 mL of 90% TBME/10% MeOH and the antidepressant compounds that eluted, were then placed in a sand bath heated to about 45° C and put under a constant flow of nitrogen for about 3 hours to evaporate to dryness. Then 0.5 mL HPLC-­‐grade methanol was added to each tube, which was vortexed until the compounds dissolved. Next, 0.25 mL of each solution was placed into a small 300 insert which was inside a regular sized amber vial. HPLC was run on each of the vials with the same conditions (70/30, 0.5 L/min). This resulted in a concentration increase of 1000 times.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.